A speech/music classifier than can distinguish whether a given audio file is a speech or any type of music. After our experiments, we were able to achieve 95% accuracy.
In order to run the script, you have to download the SpeechMusic Dataset from GTZAN website. You can also specify the audio file you want to be classified. Please make sure all the dataset and the audio files are in the same directory as the script.